Available online at www.sciencedirect.com

Journal

www.elsevier.com/locate/jterra

ScienceDirect

Journal of Terramechanics 72 (2017) 1–8

ofTerramechanics

Modeling of share/soil interaction of a horizontally reversibleplow using computational fluid dynamics

Lin Zhu a,⇑, Jia-Ru Ge a, Xi Cheng a, Shuang-Shuang Peng a, Yin-Yin Qi a, Shi-Wu Zhang b,De-Quan Zhu a

aLab of Mechanical Structure & Biomechanics, School of Engineering, Anhui Agricultural University, Hefei 230036, ChinabDepartment of Precision Machinery & Precision Instrumentation, University of Science & Technology of China, Hefei 230027, China

Received 27 May 2016; received in revised form 26 October 2016; accepted 17 February 2017Available online 21 March 2017

Abstract

The horizontally reversible plow (HRP) is currently widely used instead of the regular mold-board plow due to its high operationalperformance. Soil pressure during HRP tillage generally has adverse effects on the plow surface, especially on either the plowshare or theplow-breast. This effect eventually shortens the tool’s service life. For this reason, this investigation used a three-dimensional (3D) com-putational fluid dynamics (CFD) approach to characterize the share/soil interaction and thus assess the effects of different tillage con-ditions on the interaction. To achieve this goal, a 3D model of the plowshare was first constructed in the commercial softwareSolidWorks, and soil from Xinjiang, China, was selected and subsequently characterized as a Bingham material based on rheologicalbehaviors. Finally, 3D CFD predictions were performed using the control volume method in the commercial ANSYS code Fluent14.0 in which the pressure distributions and patterns over the share surface were addressed under different tillage speeds in the rangeof 2–8 ms�1 and at operational depths ranging from 0.1 to 0.3 m. The results show that the maximum pressure appeared at theshare-point zone of the plowshare and that the increase in soil pressure was accompanied by either higher tool speed or greater opera-tional depth. The calculated results qualitatively agreed with the preliminary experimental evidence at the same settings according toscanning electron microscopy (SEM). Once again, the CFD-based dynamic analysis in this study is demonstrated to offer great potentialfor the in-depth study of soil-tool interactions by simulating realistic soil matter.� 2017 ISTVS. Published by Elsevier Ltd. All rights reserved.

Keywords: Soil-share interaction; Computational fluid dynamics (CFD); Tillage speed; Operational depth; Horizontally reversible plow (HRP)

1. Introduction

In soil excavation processes, tillage tool interaction isthe largest energy-consuming operation because it operatesvia soil cutting with the objective of attaining suitable con-ditions for crop production (Natsis et al., 2008). The mold-board plow is widely used in tillage, and its plowshare andmold-board are the two main soil-engaging components.Tillage is the mechanical manipulation of the soil in the

http://dx.doi.org/10.1016/j.jterra.2017.02.004

0022-4898/� 2017 ISTVS. Published by Elsevier Ltd. All rights reserved.

⇑ Corresponding author.E-mail address: zl009@mail.ustc.edu.cn (L. Zhu).

tillage layer and results in severe wear on either the plow-share or mold-board (Eltom et al., 2015). The soil pressuredistribution and pattern over the plow surface are demon-strated to be strongly associated with the appearance ofabrasive wear (Natsis et al., 2008).

The horizontally reversible plow (HRP) is a novelmold-board plow developed by the Xin-Jiang AgriculturalMechanization Institute (XJAMI) of China. The uniqueadvantage of HRP, which is quite different from regularmold-board plows, is that it can facilitate continuous andalternative commuting tillage with excellent operationalperformance (i.e., steady tilling and orderly soil cutting)

2 L. Zhu et al. / Journal of Terramechanics 72 (2017) 1–8

(Zhu et al., 2006). Fig. 1 shows the commuting tillage pro-cess of HRP in real field conditions, where symbols A andB separately denote the two different limited tillage posi-tions of HRP, the pink curves indicate the hypothesizedflow of soil clods on the plow surface, and n is the rota-tional speed of the basic beam (BB) with respect to themain beam (MB) (Zhu et al., 2006). The rotational speedof the plow body is the same as that of the basic beam(BB) with respect to the main beam (MB) because it isattached to the BB.

In practical HRP tillage operations, severe abrasivewear is always found on the plow surface, especially oneither the plowshare or plow-breast. Fig. 2 shows the wornarea at the share-point of the plowshare at a tool speed of1.67 ms�1 and an operational depth of 0.27 m. This effectnot only shortens the tool service life but also affects theplow qualities due to the initial shape changes(Shmulevich, 2010). For the purpose of tool design opti-mization and performance improvement, a better under-standing of the interaction between soil and share isessential to alleviate the worn area that appears on theshare-point. In this paper, computational fluid dynamics(CFD) modeling was proposed to investigate the interac-tion between the soil and the HRP plowshare and examinethe wear phenomenon.

Currently, four major methods are available for theinvestigation of the soil-tool interactions, including empir-ical and semi-empirical methods, dimensional analysis, thefinite element method (FEM) and the distinct elementmethod (DEM) (Karmakar et al., 2007; Barker, 2008).Empirical and semi-empirical models also exist, i.e., math-ematical solutions that describe the soil-tool interactionthrough parametric studies, but they encounter limitationsin the pursuit of a better understanding of the soil-toolinteraction because they are not based on the real soil fail-ure shapes. The dimensional analysis technique is appliedto different parameters of the tillage system, but it is proneto distorted models and obscurity in two-and three-dimensional problems.

The finite element method (FEM) can overcome thedrawbacks of the previously mentioned analytical methodby supplying the soil-tool interaction with additional

Soil clod

Fig. 1. Commuting tillage process of HRP.

information (such as the progressive soil failure zone andfield of stress), but the limitation of FEM is large geometrydeformation (Barker, 2008; Karmakar, 2005; Xu et al.,2015). This problem indicates that FEM might not accu-rately simulate the complex dynamic interactions betweenthe soil and tool. As a potential new technique, the distinctelement method (DEM) can implement the particle dynam-ics interaction to obtain adequate soil reaction forces, but itfaces certain difficulties in simulating the real particle sizeand shape of soils (Shmulevich, 2010). Currently, differentauthors are working to simulate the soil-tool interactionusing DEM, but DEM is limited to simulation of wear dis-tribution on tools (Shmulevich, 2010).

Compared with the above four methods, the mainadvantage of computational fluid dynamics (CFD) is thatit can create a detailed description of the real soil flow dis-tributions (Karmakar, 2005). Hence, CFD is an alternativecomputational modeling that can assess the pressuredistributions that indicate the relative wear of the tool fromsoil-tool interactions. Considering the soil rheologicalbehavior, Karmakar and Kushwaha (2006) proposed acomputational fluid dynamics (CFD) method for soil-toolinteraction from a visco-plastic fluid flow perspective andobtained the soil stress patterns around a simple vertical til-lage tool and the pressure distributions on the tool surface.However, simulation results of the interaction between thesoil and HRP are not available in the literature.

Therefore, this study investigated the feasibility and effi-cacy of computational fluid dynamic (CFD) modeling insoil-HRP interaction. The CFD-based predictions pre-sented in this paper are primarily focused on the pressurepatterns and distributions over the HRP plowshare underhigh-speed tillage. These numerical simulations werefocused on the two key factors, i.e., tool speed and opera-tional depth, that strongly affect the HRP tillage perfor-mance (Ranjbar et al., 2013).

2. Materials and methods

2.1. Model geometry of the plowshare

The plowshare of the HRP has a curved surface, theoverall dimensions of which are 0.69 m in length, 0.12 min width and a 0.01 m maximum height (Zhu et al.,2008). These geometric parameters are strongly correlatedwith the soil types and the operating conditions, therebyaffecting tillage quality and power consumption (Gill andVanden Berg, 1967). Identification of the plowshare surfaceis, therefore, essential to better understand the soil-plowshare interaction under real field conditions. In thisstudy, combined feature-based modeling and scanningmeasurement were used to construct the 3D geometricmodel of the plowshare. The detailed procedures aredepicted as follows. First, the real plowshare surface wasscanned using a 3D touch-probe measuring bench(Explorer 07.10.05, Qindao, P.R. China) to determine theCartesian coordinates of the points in the intersection of

Fig. 2. (A) Plow-surface of HRP, (B) SEM micrograph of the worn plowshare at a tool speed of 1.67 ms�1 and an operational depth of 0.27 m.

Fig. 4. Three-dimensional model of the plowshare.

L. Zhu et al. / Journal of Terramechanics 72 (2017) 1–8 3

the vertical planes with the plow surface (Fig. 3). Second,all points obtained were reproduced and joined by a splinein the 3D design software SolidWorks. Finally, the modelgeometries of the plowshare surface were created with thefeature-based modeling approach (Fig. 4) (Zhu et al.,2016).

2.2. Three-dimensional CFD modeling

2.2.1. Hypothesis for CFD predictions

Generally, soil rheology in tilling operations is highlycomplicated, possibly due to variable deformation in thesoil-tool interaction. Soil flow is predominantly inducedby the stress difference. In other words, the soil flow is ini-tiated as the stress acting on the inter-aggregate contactexceeds the threshold stress, i.e., yield stress. In detail, ifthe applied stress is less than the yield stress, the soil flowacts as a viscous-plastic fluid, and if the stress exceeds theyield stress, the soil flows in a manner similar to a fluid(Terzaghi, 1948; Bird et al., 1983). In the foregoing process,the relationship between stress and soil flow rate is non-linear, and the soil flow pattern can be considered as a lam-inar flow in accordance with the high molecular weight

Fig. 3. Three-dimensional touch-probe measuring.

(Terzaghi, 1948; Desai and Phan, 1980; White, 1999). Forthis reason, the dynamic process of tillage can be studiedvia non-Newtonian flow behavior from a fluid flow per-spective. This approach indicates that the soil-tool interac-tion can be viewed as an external flow over a bluff body(Karmakar, 2005). To investigate the flow interaction, avolume of fluid is commonly defined for the mechanicalcharacteristics of the specific volume fraction. Addition-ally, because the volume fraction consists of soil, waterand air, a multiphase fluid flow method should be consid-ered (Barker, 2008; Terzaghi, 1948).

2.2.2. CFD-based flow model

During the soil/HRP interaction, the plowshare is con-sidered as a stationary tool located within a visco-plasticflowing domain. This domain represents the soil, i.e., the

Fig. 5. Soil flow pattern on the plowshare surface (Zhu et al., 2016).

4 L. Zhu et al. / Journal of Terramechanics 72 (2017) 1–8

plowshare is placed in a free surface domain and acts as abluff body obstruction. Fig. 5 schematically illustrates thesoil flow pattern when the plow-body is operated fromright to left. Note that this flow configuration is based onthe real tillage tool interaction. As shown in Fig. 5, the pri-mary domain of the soil flow is a surface with an area of0.0379 m2 (see the yellow1 zone in Fig. 5).

Inlet:Outlet:

L

velocity normal to

pressure; approximately half of the top

share-point zone

2.2.3. Governing equations and constitutive relations

The CFD-based numerical simulation approach is usedto investigate particle movements in a system, includingvelocity and stress distribution fields in a well-defined formto which the mass and momentum conservation equationsare applied, respectively (Desai and Phan, 1980; White,1999). Due to the effectiveness of the Navier-Stokes equa-tions for any fluid flow, the soil-tool flow interaction canalso be investigated by incorporating non-Newtonianparameters into these basic equations (Barker, 2008; Xuet al., 2015).

We apply the mass conservation through the controlvolume, and thus the continuity equation is written asfollows:

@q@t

þ @

@xiðqliÞ ¼ 0 ð1Þ

where t, q, li and xi are the flow time, fluid density, direc-tional velocity and displacement of the fluid, respectively.

The variation in the time rate of the density at any flowlocation is balanced by the net mass flux at the correspond-ing point. For simplification, if the soil is consideredincompressible and assumed as a single-phase continuousmedium with a constant density, q should be a bulk densitywith any pore water in the soil. Thus, Eq. (1) can bereduced to the following form (see Eq. (2)), which indicatesthat the volume of the differential fluid element does notchange.

@

@xiðqliÞ ¼ 0 ð2Þ

According to Newton’s second law, the relationshipbetween the acceleration of a fluid element and the netforce acting on it can be presented (Desai and Phan,1980; White, 1999):

qDli

Dt¼ qg � @P

@xiþ @sij

@xjð3Þ

where D/Dt is the function of substantial or total deriva-tive, g is the acceleration of gravity, P is the hydrostaticpressure, and sij is the shear stress tensor.

The material or substantial derivative is a function ofboth temporal and spatial changes and can be written asfollows (White, 1999):

1 For interpretation of color in Figs. 2, 5, 7, and 11, the reader is referredto the web version of this article.

Dli

Dt¼ @li

@tþ lj

@li

@xjð4Þ

Eq. (3) indicates that the acceleration of the fluid element isbalanced jointly by the gravitational force, pressure(hydrostatic stress) and viscous (hydrodynamic) stress.Therefore, it is concluded that the different aspects of thereal soil-tool dynamics interaction (such as draft force,pressure and soil failure due to visco-plastic deformation)can be characterized using the fluid flow method.

For a Bingham plastic fluid, the generalized stress-strainrate can be written as follows (Desai and Phan, 1980;White, 1999):

sii ¼ ri ¼ 2l@l@x

ð5Þ

sij ¼ sy þ l@ui@xj

þ @lj

@xi

� �¼ sy þ ly for jsijj > sy ð6Þ

c ¼ 0 for jsijj 6 sy ð7Þwhere c is the shear rate, si is the yield stress, and µ is theconstant viscosity or plastic viscosity.

Note that in this study, the soil yield stress was assumedindependent of position or orientation of the fluid particlesto satisfy the homogeneity and isotropy of the visco-plasticfluid.

2.2.4. Numerical modeling and boundary conditions

For the purpose of examining the soil flow around theplowshare, different tillage speeds, operational depths andisothermal conditions were all included in this study. Thecontrol volume method in a commercially available CFDcode, i.e., ANSYS Fluent 14.0 (ANSYS FLUENT), wasused to predict the pressure distributions and patterns overthe share-surface of the HRP. Fig. 6 illustrates the CFDgrid of the plowshare with 5399 tetrahedral cells in Gambit2.4 (ANSYS FLUENT).

During HRP tillage, the soil is first drawn into the com-putational domain through the share-point with an inletvelocity and mostly flows out from the center section ofthe plowshare (Figs. 5 and 6). Thus, the detailed boundaryconditions for the CFD simulations are specified as fol-lows: (1) The velocity component normal to the inletboundary is 2–8 m/s, (2) the pressure boundary is appliedat the outlet, (3) no slip occurs at the wall boundaries at

O

Fig. 6. CFD grid of the plowshare.

Table 1Maximum pressures over the plowshare at different operational depths (H)and different tool speeds (V).

Pmax H = 0.1 m H = 0.2 m H = 0.3 m

V = 2 ms�1 17.6 MPa 21.6 MPa 30.5 MPaV = 4 ms�1 35.3 MPa 43.4 MPa 61.4 MPaV = 6 ms�1 53.2 MPa 65.3 MPa 92.7 MPaV = 8 ms�1 71.3 MPa 87.4 MPa 124 MPa

L. Zhu et al. / Journal of Terramechanics 72 (2017) 1–8 5

the bottom or sides of the plowshare, and (4) the free-surface grid movement is confined in the 3D surface region.In addition, the different operational depths of HRP, i.e.,H = 0.1–0.3 m, were also considered in the computationaldomain. For all calculations, the numerical results werebased on the criterion that the residual of each equationshould be smaller than 1.0 � 10�5.

2.2.5. Soil parameters for CFD prediction

For real soil with water and air, as depicted previously,the multiphase fluid flow should be considered. However,for simplicity, a single-phase laminar flow was used tocharacterize the dynamic interaction between the soil andplowshare in this study. Currently, the HRP is widely usedin Xin-Jiang in China. According to the corresponding soilproperties (a bulk density of 1453 kg m�3, a cone indexvalue of 398 kPa and a moisture content of 17%), the soilpresented in this work was characterized as an incompress-ible, isotropic and homogeneous Bingham material (Gilland Vanden Berg, 1967). Thus, the dynamic parametersof the soil can be determined using the strain-rate basedsoil torsional rheometer developed by XJAMI (Zhuet al., 2008). The measured dynamic parameters of the soilfor CFD modeling were a yield stress of 15.5 kPa and anapparent viscosity of 167 kPa s. Note that during the tests,two independent variables that influence the yield stressand viscosity, i.e., soil moisture content and cone index,respectively, were considered.

3. Results and discussion

According to Ranjbar et al. (2013), Terzaghi (1948), thesoil pressure distribution and pattern during tillage havesignificant effects on the wear phenomena of the plowsharesurface and hence on the tool life. Both tool speed andoperational depth are two important tillage conditions.The following CFD-based predictions, therefore, focusedon pressure patterns and distributions over the plowshareunder the different tool speeds and operational depths.

Fig. 7 shows the CFD-based prediction under differentoperational depths, i.e., 0.1, 0.2, and 0.3 m and differenttool speeds, i.e., 2, 4, 6, and 8 ms�1. The red, green andblue colors in Fig. 7 respectively represent the maximum,median and minimum pressure sections over the plowsharesurface. The results showed that for the pressure distribu-tion and pattern over the plowshare, reasonably good

Fig. 7. CFD-based prediction of the plowshare at different operationaldepths and various tool speeds.

qualitative agreement was observed between the two tillageconditions, i.e., tool speed and operational depth. In otherwords, the maximum pressures both appeared at the share-point of the plowshare. However, a discrepancy in magni-tude was noted for the maximum pressure. The detailedvalues are listed in Table 1.

Fig. 8 plots the pressure variations over the plowsharewith changes in distance at the different operational depths(H), i.e., 0.1, 0.2 and 0.3 m. The distance (D) presented inthis work is the length between any point, e.g., point Aon the normal L of the share-point section and the share-point O, as shown in Figs. 6 and 9.

As observed, for the range of tool speeds from 2 to8 ms�1, a greater operational depth was accompanied byhigher pressure at the plowshare, and the maximum pres-sures all appeared at the share-point section, especiallyclose to the share-point. At operational depths rangingfrom 0.1 m to 0.3 m, the maximum pressure approximatelyincreased from 17, 34, 52, 70 MPa to 30, 60, 90, and120 MPa, respectively, corresponding to tool speeds of 2,4, 6, and 8 ms�1. Furthermore, for any tool speed andoperational depth, as shown in Fig. 8, the highest pressureat the share-point gradually decreased with the distancefrom the share-point.

Fig. 10 shows the pressure variations on the plowsharewith distance under different tillage speeds, i.e., 2, 4, 6and 8 ms�1. Note that the distance presented has the samemeaning as in Fig. 8. As observed, at operational depthsranging from 0.1 m to 0.3 m, greater tool speed alsoresulted in higher pressure on the plowshare. The maxi-mum pressures on the plowshare all appeared at theshare-point. For tool speeds in the range of 2–8 ms�1, thehighest pressures approximately increased from 17, 21,and 30 MPa to 70, 85, and 120 MPa, respectively, corre-sponding to operational depths of 0.1, 0.2 and 0.3 m. Inaddition, similar to in the information shown in Fig. 8,for any tool speed and operational depth, the variation ten-dency of the pressures on the plowshare was that the high-est pressure at the share-point also gradually decreasedwith the distance from the share-point.

It can be concluded from Figs. 7, 8 and 10 that for dif-ferent tillage speeds and operational depths, the pressureon the plowshare increases with either tillage speed or oper-ational depth and that the maximum pressures on theplowshare all appear at the share-point zone. This observa-tion might be attributed to the fact that the share-point isthe main soil engagement section of the plowshare, where

Fig. 8. Pressure variation on the share-point with change in distance at different operational depths (H) and a constant tillage speed of (A) 2 ms�1, (B)4 ms�1, (C) 6 ms�1 and (D) 8 ms�1.

Fig. 9. Distance between any point on the normal of the share-pointsection and the share-point of the plowshare.

6 L. Zhu et al. / Journal of Terramechanics 72 (2017) 1–8

a large amount of energy is necessary to cut, break downand reduce the clod size. This action eventually leads tothe greatest stress concentration in the soil surroundingthe share-point zone.

Generally, the soil front propagation or soil disturbancezone in the tilling operation can be interpreted via the

pressure distribution on the tool surface (Terzaghi, 1948;Anonymous, 2001; Adichi and Yoshioka, 1973). Thus,the soil plastic flow section can be considered as the soildisturbance zone with particular soil stresses. Conse-quently, the maximum soil stress in the flow domain indi-cates the highest pressure on the tool surface. The share-point is the main soil engagement zone of the plowshare.According to Coulomb’s passive earth pressure theory, asuccession of the shear planes leads to block separationof the soil mass, which in turn causes shear failure to initi-ate from the tool cutting edge (Anonymous, 2001). It is,therefore, inferred that the maximum load on the plow-share due to soil-tool interaction lies at the share-pointzone, as numerically demonstrated by CFD modeling inFigs. 7, 8 and 10, and thus supports the assumptions ofCoulomb’s passive earth pressure theory.

From a fluid flow perspective, because the soil in thisstudy was modeled using the Bingham constitutive law, acharacteristic peculiarity concerning the fluidity of the

Fig. 10. Pressure variation on the plowshare with change in distance at the different tillage speeds (V) and a constant operation depth of (A) 0.1 m, (B)0.2 m and (C) 0.3 m.

L. Zhu et al. / Journal of Terramechanics 72 (2017) 1–8 7

visco-plastic medium appears at the boundaries betweenthe tool and soil. According to Adichi and Yoshioka(1973), these boundaries can divide the flow fields into fluidregions and rigid regions. The fluid or plastic flow regionsare located close to the boundary of the flow, where thepressure gradient is rather high. Hence, the maximum pres-sure at the share-point zone of the plowshare might becaused by the stagnation point of fluid flow.

4. Comparison of numerical and preliminary measurements

The CFD-based simulations in this study showed thatall of the maximum pressures appeared at the share-pointzone of the plowshare. The preliminary measurementsfrom SEM showed that at a tool speed of 1.67 ms�1 andoperational depth of 0.27 m, the most severe wear areaon the plowshare also occurred at the share-point section(Fig. 2). This location is where the maximum pressure onthe plowshare lies, as shown in Figs. 7, 8 and 10. At thesame time, the CFD-based predictions in this study furthershowed that the maximum pressure at the share-point zonewas increased with either tillage speed or operationaldepth. Under similar tillage conditions, for example, whenthe operational depth was increased from 0.2 m to 0.3 m ata constant tool speed of 2 ms�1, the maximum pressure at

Fig. 11. Comparison between CFD predictions an

the share-point was increased from 21.6 MPa to 30.5 MPa(Table 1, Fig. 11(A) and (B)). The recent experimental mea-surements also validated that at a tool speed of 1.67 ms�1,a more severely worn area (compared to Fig. 2) appearedat the share-point when the operational depth wasincreased from 0.27 m to 0.36 m (see the red arrows inFigs. 2 and 11(C)). The highest pressure incurs the mostsevere wear on the tool surface in the soil-cutting process(Koolen, 1983). Based on qualitative comparisons amongFigs. 2 and Figs. 7 and 11, we infer that the most severewear over the plowshare surface under other tool speedsand operational depths should also appear at the share-point section and might even be more severe than thatshown in Fig. 2.

In general, the overall trend of the numerical resultsshows good qualitative agreement with the measurements.This outcome once again demonstrates that CFD modelingcan be feasibly and effectively used to study soil-tool inter-action and hence tool design improvement. However, thecurrent results suggest further investigations. First, the ade-quate values of the more severe wear areas on the plow-share under the different tool speeds and operationaldepths were numerically determined using the maximumpressure profiles obtained in this study. Second, additionalexperimental work is also required to investigate the wear

d the experiment at similar tillage conditions.

8 L. Zhu et al. / Journal of Terramechanics 72 (2017) 1–8

phenomenon on the plowshare, including pressure distribu-tion measurements and SEM-based analyses under theidentified operation conditions. Higher pressure indicatesa stronger soil-tool interaction and thus leads to more sev-ere wear and shorter tool service life.

5. Conclusions

Based on the merits and weaknesses of the existing soil-tool modeling methods, 3D CFD predictions were pro-posed to investigate the flow interaction between the soiland plowshare during HRP high-speed tillage. This studyfocused on the pressure patterns and distributionsover the plowshare under different tillage speeds, i.e.,2–8 ms�1, and operational depths, i.e., 0.1–0.3 m. Thenumerical simulations demonstrated that the maximumpressures all appeared at the share-point zone of the plow-share and that the pressures at the share-point increasedwith either tillage speed or operational depth. Preliminaryand further measurements both showed that the mostsevere wear areas appeared at the share-point section ofthe plowshare and that the wear areas at the share-pointbecame more severe with increasing operational depth.Because the highest pressure is strongly associated withthe most severe wear phenomenon for the plowshare, theCFD-based predictions in this study show good qualitativeagreement with the preliminarily measured evidence fromSEM. CFD-based modeling in the tillage area is once againdemonstrated to be feasible and effective for the in-depthstudy of soil-tool interactions by simulating realistic soilmanners. The research results in this study offer a newdimension for agricultural tool design and especially forHRP design.

Acknowledgments

The authors acknowledge financial support for thiswork from the National Natural Science Foundation ofChina (Grant No. 51575003), the Anhui Provincial NaturalScience Foundation (Grant No. 1508085ME71) and theKey Project of Anhui Education Committee (Grant No.KJ2015A031). The authors also thank the reviewers fortheir valuable input.

References

Adichi, K., Yoshioka, N., 1973. On creeping flow of a visco-plastic fluidpast a circular cylinder. Chem. Eng. Sci. 28 (1), 215–226.

Anonymous, 2001. Earth pressure theory and applications. In: Trenchingand Shoring Manual. California Department of Transportation,Sacramento, CA, USA.

Barker, M.E., 2008. Predicting Loads on Ground Engaging Tillage ToolsUsing Computational Fluid Dynamics. PhD Thesis. Iowa StateUniversity, Iowa, USA.

Bird, R.B., Dai, G.C., Yarusso, B.J., 1983. The rheology and flow ofvisco-plastic materials. Rev. Chem. Eng. 529 (1), 1–70.

Desai, C.S., Phan, H.V., 1980. Computational methods in non-linearmechanics. North Holland Publishing Company, 205–245.

Eltom, A.E. Farid, Ding, Weimin, Ding, Qishuo, Ali, Abu baker B.,Adam, B. Eisa, 2015. Effects of trash board on moldboard plowperformance at low speed and under two straw conditions. J.Terramch. 59, 27–34.

Gill, W.R., Vanden Berg, G.E., 1967. Soil dynamics in tillage and traction.In: Agriculture Handbook No. 316. Agricultural Research Service, U.S. Department of Agriculture, 511.

Karmakar, S., 2005. Numerical Modeling of Soil Flow and PressureDistribution on a Simple Tillage Tool Using Computational FluidDynamics. PhD Thesis. University of Saskatchewan, Canada.

Karmakar, S., Kushwaha, R.L., 2006. Dynamic modeling of soil–toolinteraction: an overview from a fluid flow perspective. J. Terrramech.43, 411–425.

Karmakar, S., Kushwaha, R.L., Lague, C., 2007. Numerical modeling ofsoil stress and pressure distribution on a flat tillage tool usingcomputational fluid dynamics. Biosyst. Eng. 97, 407–414.

Koolen, A.J., 1983. Agricultural Soil Mechanics. Advanced Series inAgricultural Sciences. Springer-Verlag, Berlin.

Natsis, A., Petropoulos, G., Pandazaras, C., 2008. Influence of local soilconditions on moldboard plowshare abrasive wear. Tribol. Int. 41,151–157.

Ranjbar, I., Rashidi, M., Najjarzadeh, I., Niazkhani, A., Niyazadeh, M.,2013. Modeling of moldboard plow draft force based on tillage depthand operation speed. Middle-East J. Sci. Res. 17, 891–897.

Shmulevich, I., 2010. State of the art modeling of soil–tillage interactionusing discrete element method. Soil Tillage Res. 111, 41–53.

Terzaghi, K., 1948. Theoretical Soil Mechanics. Chapman and Hall,London.

White, F.M., 1999. Fluid Mechanics, 4th ed. WCB/McGraw-Hill,Columbus, OH, USA.

Xu, Lichao, Zhang, Shiwu, Jiang, Nan, Xu, Ronald X., 2015. A hybridforce model to estimate the dynamics of curved legs in granularmaterial. J. Terrramech. 59, 59–70.

Zhu, Lin, Yin, Chen-Long, Chen, Fa, 2006. The application of virtualprototype technology on analyzing the kinetic and dynamic charac-teristic of horizontally reversible plow. Int. J. Jpn. Soc. Mech. Eng.Ser. C 41, 247–252.

Zhu, Lin, Xu, Jin-Liang, Yin, Cheng-Long, Kong, Fan-Rang, Kong,Xiao-Ling, 2008. Design of plowshare of horizontal reversible plowwith computer simulation technique. J. Syst. Simul. 20 (13), 3455–3458.

Zhu, Lin, Peng, Shuang-Shuang, Xi-Cheng, Qi, Yin-Yin, Zhang, Wen-Feng, Xu, Liang-Yuan, 2016. Virtual assembly geometric semanticsand constraint for an improved three-dimensional model of horizon-tally reversible plow. Int. J. Mech. Sci. Technol. 30 (1), 257–266.

Zhu, Lin, Peng, Shuang-Shuang, Cheng, Xi, Qi, Yin-Yin, Ge, Jia-Ru, Yin,Cheng-Long, Jen, Tien-Chien, 2016. Combined finite element andmulti-body dynamics analysis of effects of hydraulic cylinder move-ment on plowshare of Horizontally Reversible Plow. Soil Tillage Res.163, 168–175.